Threat actors have begun actively exploiting a recently patched, critical vulnerability in Cisco Expressway to compromise enterprise networks. The intrusions, which occurred shortly after the public availability of a proof-of-concept exploit, allowed attackers to execute operating system commands as root and pivot from external network boundaries into protected internal infrastructure.

The campaign represents the first confirmed case of in-the-wild exploitation targeting the flaw, according to a report from Russian cybersecurity firm F6 covered by The Hacker News. The attackers compromised an organization by initially striking internet-facing appliances before moving deeper into the victim's environment.

What Happened in the Attack

According to telemetry analyzed by F6, the threat actors targeted vulnerable Cisco Expressway devices running version X15.5.3. The attackers successfully compromised two internet-exposed Cisco Expressway systems. Using one of these initial beachheads, the adversaries successfully gained access to two additional cluster nodes situated entirely within the organization's internal network.

Once armed with root privileges across the cluster, the unidentified intruders launched reconnaissance and lateral movement operations. The attackers deployed custom Python scripts to sweep the internal network, map accessible SMB shares, communicate with SQL Server databases, and establish connections via Windows Remote Management (WinRM). The activity culminated in accessing internal GitLab repositories to harvest credentials and exfiltrate sensitive enterprise data.

F6 noted that analysis of the custom Python tooling suggested the scripts were likely crafted with the help of artificial intelligence tools, evidenced by inline comments and detailed descriptions of code fixes embedded directly within the files.

Technical Details of CVE-2026-76466

The vulnerability at the center of the intrusion is tracked as CVE-2026-76466 and carries a near-maximum CVSS severity rating of 9.8. It allows an unauthenticated, remote adversary with network access to execute arbitrary commands on the underlying operating system with root-level permissions.

In its security advisory, Cisco explained that the flaw stems from the insecure handling of persistent cookies combined with flawed authentication mechanisms in clustered deployments. Specifically, the system fails to sufficiently validate client certificates even when the TLS Verify Mode configuration is set to Enabled for clustering.

An attacker possessing the persistent cookies can leverage this lack of validation to connect directly to the local database service. From there, the attacker can invoke specific remote procedure call (RPC) commands, triggering arbitrary command execution with root privileges on the targeted device.

Affected Versions and Vendor Response

Cisco addressed the security defect on October 21, 2026, delivering software updates across affected branches. The vulnerability impacts multiple releases of the Cisco Expressway Series:

  • Cisco Expressway Series Release X14 and earlier: Administrators must migrate to a supported, fixed release.
  • Cisco Expressway Series Release X15.5.3 and earlier: Customers must upgrade to fixed version X15.5.20.

The activity observed by F6 has not yet been linked to a specific threat actor or established hacking group. However, researchers warned that perimeter devices remain prime targets for adversaries looking to bypass border protections and establish reliable access to core infrastructure.

What You Should Do

Security teams managing Cisco Expressway deployments should prioritize defensive measures immediately to counter ongoing exploitation:

  • Apply available vendor updates: Upgrade all Cisco Expressway appliances to version X15.5.20 or an equivalent fixed release as directed by Cisco's advisory.
  • Inspect Erlang system logs: Audit the erlang.log file on all Expressway devices for anomalous entries, specifically checking for the presence of the string [email protected], which indicates potential exploit activity.
  • Search for unauthorized scripts: Check system storage paths for suspicious Python tools, specifically looking for files matching the pattern /mnt/harddisk/i{0-9}{4}[a-zA-Z]?/*.py.
  • Block known adversary infrastructure: Monitor and restrict external network traffic associated with the observed malicious IP addresses: 66.179.31[.]32 and 18.130.47[.]116.
  • Review network segmentation: Enforce strict network zoning between edge communications appliances and internal resources, restricting lateral protocols such as WinRM, SMB, and database access from perimeter clusters.